Stability of the retinal image under normal viewing conditions and the implications for neural adaptation.

Previous studies have demonstrated that the visual system adapts to the specific aberration pattern of an individual's eye. Alterations to this pattern can lead to reduced visual performance, even when the Root Mean Square (RMS) of the wavefront error remains constant. However, it is well-established that ocular aberrations are dynamic and can change with factors such as pupil size and accommodation. This raises an intriguing question: can the neural system adapt to continuously changing aberration patterns? To address this question, we measured the ocular aberrations in four subjects under various natural viewing conditions, which included changes in accommodative state and pupil size. We subsequently computed the associated Point Spread Functions (PSFs). For each subject, we examined the stability in the orientation of the PSFs and analyzed the cross-correlation between different PSFs. These findings were then compared to the characteristics of a distribution featuring PSF shapes akin to random variations. Our results indicate that the changes observed in the PSFs are not substantial enough to produce a PSF shape distribution resembling random variations. This lends support to the notion that neural adaptation is indeed a viable mechanism even in response to continuously changing aberration patterns.